Design and fabrication of a multi-stationary phase thin-layer chromatography system for the rapid chemometric fingerprinting of B. balsamifera and V. negundo

Date of Award

2018

Document Type

Thesis

Degree Name

Master of Science in Chemistry

Department

Chemistry

First Advisor

Enriquez, Erwin P., Ph.D.

Abstract

Despite rising consumer demand for herbal medicinal products in the Philippines, the growth in the industry has been limited by the supply of quality raw plant materials for processing into finished products. One of the major factors contributing to this limited supply is the lack of accessible technology for quality control at the farmer-supplier level, which leads to high rates of supply rejection. This study aimed to address the lack of accessible quality assessment tools by developing a field-ready thin-layer chromatography (TLC) method coupled with chemometric fingerprinting that can identify common concerns such as incorrect plant variety, contamination, plant quality degradation, or sample mishandling or mislabeling. To ensure the adoptability of the method for field use, a novel multi-stationary phase(MSP) TLC plate was designed and fabricated for the one-step multiple development of ethanolic plant extracts with a 60:40 (v/v) ethanol-water mobile phase. This approach minimizes the use of volatile organic solvents typically used in TLC protocols, while varying the selectivity of chromatographic separation by using multiple phenylandoctyl-modified stationary phases printed on a silica gel TLC plate. The method was applied to B. balsamifera and V. negundo ethanolic extracts as model systems, and was coupled with image analysis and one-class soft independent modeling of class analogy (SIMCA) to classify samples as within-specifications (WS) or off-specifications(OS) based on their MSP-TLC profiles. WS samples used in this study were pure plant samples of known processing history that passed the Pharmacopeia acceptance criteria, whereas OS samples used in this study were samples that did not pass the acceptance criteria and/or were intentionally degraded or contaminated. The optimized SIMCA model for B. balsamifera demonstrated 75.0% and 95.7% sensitivity and specificity, respectively. The increased specificity of the method is an improvement over the Pharmacopeia method which is subject to misclassification of adulterations as high as 50% (w/w). The optimized SIMCA model for V. negundo, meanwhile, demonstrated 83.3% and 78.6% sensitivity and specificity, respectively. However, the model could only correctly detect contaminations as low as 30% (w/w), which is nonetheless a significant improvement over the Pharmacopeia method which can potentially misclassify samples up to 80% (w/w) adulteration. The lower sensitivity of the classification models, on the other hand, may be due to the greater responsiveness of the method to unusual features in the profiles compared to the Pharmacopeia acceptance criteria. Overall, this study serves as a proof-of-concept demonstration of the use of a general one-step multiple development TLC method that when coupled with image analysis and chemometric fingerprinting, has greater specificity with respect to contaminations compared to the standard Pharmacopeia method. The proposed method can serve as a promising field-ready quality assessment tool that is user-friendly, robust, portable, reliable, and safe for end-users such as farmers and processors.

Comments

The C4.S537 2018

Link to Full Text

Share

COinS